how unpack function will work in perl for this code $str =~ s/([^\w ])/'%'.unpack('H2', $1)/eg; - perl

Question

i have a code in perl $str =~ s/([^\w ])/'%'.unpack('H2', $1)/eg; i am not undestanding what value will be stored in $str

Answer 1

Assuming $str is encoded using UTF-8, and assuming the code you provided is followed by $str =~ s/ /+/g, the result is a url-encoded string safe for use in URLs.
Specifically, the line of code in question replaces every non-word except spaces with a three character sequence starting with % and followed by two hex digits representing the character number.
For example,
foo's ⇒ foo%27s
20% ⇒ 20%25
A better solution would be to use uri_escape (for strings encoded using UTF-8) or uri_escape_utf8 (for strings of Unicode Code Points aka decoded strings) from URI::Escape.

Answer 2

Provided line of code modifies $str value according substitute rule set s/([^\w ])/'%'.unpack('H2', $1)/eg.
How does it work:
[^\w] - look at $str for character not \w known as complement to \w
\w - represents range [A-za-z0-9_], punctuation chars and Unicode marks see perlre
([^\w]) capture found character, 'store' it in $1
regex modifier e evaluates '%'.unpack('H2',$1) as substitution string
unpack('H2',$1) - unpack $1 with template 'H2' (hex representation of byte associated with $1)
take '%' and concatenate it with unpacked result
use result from step 6 as replacement string
regex modifier g instructs to make this operation for all occurrences in the $str
Without knowing initial $str value before this operation, impossible to evaluate final result.
If initial value is known then you can evaluate result by visiting https://regex101.com/ website.
Nothing could speak louder than sample code demonstrating transformation
use feature 'say';
$msg = "Date: Mar 6 2020, Msg: soon Alex's birthday";
$msg =~ s/([^\w ])/'%'.unpack('H2', $1)/eg;
say $msg;
Output
Date%3a Mar 6 2020%2c Msg%3a soon Alex%27s birthday
Following code demonstrates how "Hello World\n" will look as hex representation (for Dada).
use feature 'say';
my $msg = "Hello World!\n";
print $msg;
my $a = unpack('H*',$msg);
say $a;
Output
Hello World!
48656c6c6f20576f726c64210a

Answer 3

You could start by trying it out and seeing if that gives you a hint.
$ perl -E'$str = "&*("; $str =~ s/([^\w ])/"%".unpack('H2', $1)/eg; say $str'
%26%2a%28
So, we have a substitution operator that looks like this:
s/PATTERN/REPLACEMENT/OPTIONS
Our pattern is ([^\w ]) which means "match every individual character that isn't a 'word character' or a space and capture that character in $1.
The replacement string is "%".unpack('H2', $1). Which means "the character '%' followed by the result of running unpack('H2', $1). unpack() here is being used to convert characters to the hexadecimal equivalent of their ASCII code. "H" means "convert to hex" and "2" means produce two hex digits".
The options are /e which means "run this code and use the output as the replacement string" and /g which means "do this for every match in the input string".
Putting that all together, you have code that:
Looks for non-word characters
Converts them to their hexadecimal escape code
Replaces them in the string
Using URI::Escape is probably a better approach.